Multiscale Nature of Stress Corrosion Cracking

G. S. Was

University of Michigan

Stress corrosion cracking (SCC) is an inherently multiscale process that requires a

combination of experiment and simulation to understand. This talk will focus on the
nature of stress corrosion cracking from atomistic to macroscopic length scales. Emphasis
will be placed on the phenomenon of irradiation assisted stress corrosion cracking
(IASCC) in austenitic stainless steels as an example of a process that can be understood
only through combinations of techniques over a range of length scales. IASCC is believed
to be driven by the combined action of localized deformation, grain boundaries and an
aggressive environment. Using novel experimental techniques employing an in-situ
capability, pre-determined and pre-characterized geometries for dislocation channel- grain
boundary interaction, and molecular statics and molecular dynamics modeling of channel-
boundary interaction, we are determining the basic processes and mechanisms by which
IG cracking initiates in irradiated austenitic Fe-Cr-Ni alloys conducted in aggressive
media at high temperature. A key aspect of this effort is that it embraces the essence of
mesoscale science, providing the linkage from atomistic simulations of dislocation
responses to the accommodation and emission of dislocations from grain boundaries, and
direct observation of dislocations with irradiation defects and grain boundaries to
macroscale experiments. A consequence of this combined effort has been the
identification of the factors most likely responsible for establishing not only the local
stress state at grain boundaries prone to failure but their location with respect to the
macroscopic applied stress.
 Multiscale simulations of stress corrosion effects and surface response to charging

J.-M. Albina1,2,3 Y. Umeno3, G. Moras1, C. Elsässer1,2, P. Gumbsch1,2

1
Fraunhofer Institute for Mechanics of Materials IWM, W¨ohlerstr. 11, 79108 Freiburg,
Germany
2
Karlsruhe Institute of Technology KIT, Institute for Applied Materials, Kaiserstr. 12,
76131 Karlsruhe, Germany
3
Institute of Industrial Science, The University of Tokyo, 4-6-1 Komaba, Meguro-ku,
Tokyo 153-8505, Japan

Stress corrosion, tribochemical wear or the mechanical response of surfaces to charging

are phenomena which require quantum mechanical treatment of the elementary physical
phenomena but at the same time reveal their macroscopic consequences only by
integration or interpretation on macroscopic scales.
As examples, I will address the dissociative chemisorption of H2 molecules at the edges
of hydrogen-induced platelet defects in silicon crystals which stabilizes the breaking up
of highly strained silicon bonds. The resulting stress-corrosion fracture process yields
atomically smooth cleavage of Si crystals along otherwise unstable (100) planes. Similar
chemical reactions have a big influence on the tribological behaviour of diamond. Linear
carbon chains that form during sliding of diamond surfaces undergo a fast degradation
process upon exposure to air. This oxidative wear process is triggered by the
chemisorption of O2 molecules that lead to the splitting and to the shortening of such sp-
hybridized structures.
Direct electrical charging of metal surfaces alters the surface properties significantly.
Electronic structure calculations allow assessing the effect of tangential strain on the
work function and the change in surface stress as a function of charging. This is
demonstrated on the 4d series of transition metals. They mostly follow a parabolic
behavior, which is dependent on the d-band occupancy until we reach the nobel metals,
which show an s-band occupation dependency. The variation of the stress-charge
coefficients is interpreted in terms of a bonding-antibonding orbital occupation.
 Multiscale modelling of materials chemomechanics:
from catastrophic brittle fracture to stress corrosion cracking

James Kermode, Giovanni Peralta, Zhenwei Li,

Marco Caccin, Federico Bianchini, Alessandro De Vita

King’s College London, Department of Physics, Strand, London, WC2R 2LS, United
Kingdom

Fracture of oxides and semiconductors is the dominant failure process underlying many materials
reliability issues, and is also relevant to mining, where huge amounts of energy are required to
fragment thousands of tons of rock into fine powder raising ever more serious sustainability
issues. At the same time, fracture remains one of the most challenging ‘multi-scale’ modelling
problems: since crack propagation is driven by the concentration of a long-range stress field at an
atomically sharp crack tip, an accurate description of the chemical processes occurring in the near
tip region is essential, as is the inclusion of a much larger region in the model systems. Both these
requirements can be met by combining a quantum mechanical description of the crack tip with a
classical atomistic model that captures the long-range elastic behaviour of the surrounding crystal
matrix, using a QM/MM (quantum mechanics/molecular mechanics) approach such as the `Learn
on the Fly’ (LOTF) scheme [1], or more recently a novel information-efficient Machine Learning
extension of LOTF, where the results of all QM calculations are stored in a database and used this
to predict forces on new configurations whenever possible, thus calling upon QM only when it is
strictly necessary [2].
Examples of the application of these techniques to ‘chemomechanical’ processes, where complex
and interrelated chemical and mechanical processes that originate at the atomic scale go on to
determine the mechanical behaviour of a material, include low-speed dynamical fracture
instabilities in silicon [3], interactions between moving cracks and material defects such as
dislocations or impurities [4], very slow crack propagation via kink formation and migration, and
chemically activated fracture, where cracks advance under the concerted action of stress and
corrosion by chemical species such as oxygen or water [5]. The mechanism underlying stress-
corrosion cracking in silica forms the topic of ongoing work, enabled by an INCITE award from
the US DoE and hosted at the Argonne Leadership Computing Facility [6]. An expansion of the
modeling approach to metals is also underway, focusing on the mechanical properties of nickel-
based superalloys, a class of materials that exhibit excellent strength and creep-resistance at high
temperatures making them suitable for the construction of efficient turbines for energy generation
and aerospace applications [7].

[1] G. Csányi, T. Albaret, M. Payne and A. De Vita, Phys. Rev. Lett. 93, 175503 (2004).
[2] Z. Li, J. R. Kermode and A. De Vita, In prep. (2014).
[3] J. R. Kermode et al., Nature 455, 1224-1227 (2008).
[4] J. R. Kermode, et al., Nat. Commun. 4, 2441 (2013).
[5] A. Gleizer, G. Peralta, J. R. Kermode, A. De Vita and D. Sherman, Phys. Rev. Lett. (accepted
Feb 2014).
[6] J. R. Kermode, O. A. von Lilienfeld and A. De Vita, SiO2 Fracture: Chemomechanics with a
Machine Learning Hybrid QM/MM Scheme. INCITE 2014 application (awarded Nov 2013).
[7] J. R. Kermode, F. Bianchini, A. Comisso and A. De Vita, Multiscale Atomistic Simulation of
the Mechanical Behaviour of Nickel-based Superalloys, 8th PRACE Project Access Call
(awarded Feb 2014).
 Self-Healing Materials and Damage from Shock Induced Nanobubble Collapse:
Reactive Molecular Dynamics Simulations

Priya Vashishta, Adarsh Shekhar, Ken-ichi Nomura, Rajiv K. Kalia, Aiichiro Nakano,
Kohei Shimamura, Fuyuki Shimojo

Collaboratory for Advanced Computing and Simulations

Departments of Chemical Engineering & Materials Science, Physics & Astronomy, and
Computer Science, University of Southern California, Los Angeles, CA 90089-0242, USA

Bonding and Structure of Ceramic-Ceramic Interfaces -

Interfaces in multicomponent systems are critical in
determining their material behavior. High-resolution TEM
identified rich interfacial phases of thickness ~ 10 Å, which
have distinct atomistic structures that do not exist in bulk
phases. An archetypal ceramic/ceramic interface is formed
Figure: Purification of between Al2O3 and SiC, which is important for broad
interfacial bonds by annealing. applications such as matrix-nanoparticle composites, high-
Total bond overlap populations
of the C- and Si-interfaces are
power electronic devices, and self-healing nanocomposites that
shown for both relaxed and autonomously heal cracks and recover strengths. For better
annealed configurations. Bonds understanding and design of Al2O3/SiC interfaces, the central
with O(red) and Al(green) are question is: What are the nature of interfacial bonding at the
shown. level and its consequence on interphase structures?
atomistic
Our quantum molecular dynamics (QMD) simulations of -Al2O3 (0001)/3C-SiC (111) interfaces
revealed profound effects of thermal annealing for producing strong interfaces consisting solely
of cation-anion bonds. A Si-terminated SiC surface and Al2O3 form a stronger interface (Si-
interface) with a Si-O bond density of 12.2 nm2, whereas the C-interface has an Al-C bond
density of 9.46 nm2. The interfacial bond strengthening is accompanied by the formation of an
Al2O3 interphase with a thickness of 2-8 Å.
Damage on Silica Surface from Shock Induced Nanobubble Collapse - Despite a great deal of
research on cavitation bubbles, several important questions about the mechano-chemistry of
bubble collapse near a solid surface remain unanswered. In particular, an atomistic understanding
is lacking. We have addressed these questions using billion-atom reactive molecular dynamics
(RMD) simulations. These are the largest RMD simulations done on 163,840 processors IBM
BlueGene/P at Argonne Leadership Computing Facility. The simulations revealed more chemical
activity in water nanojets and vortex rings formed by collapsing nanobubbles, in cavitation pits
on silica surface and around secondary shock waves. We found that the structural damage is
intimately related to the chemical processes initiated by water nanojets, and can be mitigated by
incorporating suitable gas in the nanobubble.
We examine chemical and mechanical damages due to shock-induced collapse of a nanobubble
near a silica slab. We also performed two sets of simulations in which the nanobubble contained
an inert gas at a density of 0.24 g/cc. The role of gas density in the stability of a single bubble is
usually examined for gas densities around 0.5 g/cc. The simulated systems contained 109 and 108
atoms and the initial bubble radii in these systems were again 97.6 nm and 40 nm, respectively.
Acknowledgement: Research supported by DOE-BES Grant Number DE-FG02-04ER46130.
[1] “Bonding and Structure of Ceramic-Ceramic Interfaces”, Kohei Shimamura,Fuyuki Shimojo,1,2 Rajiv
K. Kalia,1 Aiichiro Nakano,1 and Priya Vashishta, Physical Review Letters 111, 066103 (2013).
[2] “Nanobubble Collapse on a Silica Surface in Water: Billion-Atom Reactive Molecular Dynamics
Simulations”, Adarsh Shekhar, Ken-ichi Nomura, Rajiv K. Kalia, Aiichiro Nakano, and Priya Vashishta,
Physical Review Letters, 111, 184503 (2013).
 Mapping Strain Rate Dependence of Dislocation-Defect Interactions in Zirconium

Yue Fan1, Sidney Yip1,2, Bilge Yildiz1

1
Department of Nuclear Science and Engineering, MIT,
Cambridge, Massachusetts 02139
2
Department of Materials Science and Engineering, MIT,
Cambridge, Massachusetts 02139

Probing the mechanisms of defect-defect interactions at strain rates lower than 106 s-1 is
an unresolved challenge to date to molecular dynamics (MD) techniques. Here we
propose a novel atomistic approach based on transition state theory and the concept of a
strain-dependent effective activation barrier that is capable of simulating the kinetics of
dislocation-defect interactions at virtually any strain rate, exemplified within 10-7 to 107 s-
1
. In particular, we investigate the interaction between an edge type dislocation and an
irradiation induced self interstitial atom (SIA) cluster in Zr as a model system.

The resulting interaction paths, energy barriers and critical stresses are found to depend
significantly on the strain rates and temperatures considered. We found that, at low strain
rate and high temperature conditions, the SIA cluster is eventually absorbed by the
dislocation forming a jog structure, and thus, assists the climb. However, at high strain
rate and low temperature conditions, the dislocation passes through the cluster. Both the
dislocation and the SIA cluster fully recover their original structures, and this process has
no contribution to climb. Our predictions of a crossover from a defect recovery at the low
strain rate regime to defect absorption behavior in the high strain-rate regime are
validated against independent, direct MD simulations at 105 to 107 s-1. Implications of the
present approach for probing molecular-level mechanisms in strain-rate regimes
previously considered inaccessible to atomistic simulations are discussed.
 Measurements and Observations on Adhesion to Brittle Materials

Rajan Tandon, Michael E. Chandross

Materials Science and Engineering Center, Sandia National Laboratories, Albuquerque,

NM 87185

Bonded systems between different material families (metals, ceramics, and polymers),
and within each family are widely used. A few examples include: polymer coated
dielectrics, encapsulated electronic packages, brazed and soldered assemblies, glass to
metal seals, plated and coated electrical contacts, coatings on tribological materials, and
arc-sprayed components. The performance of such multi-materials systems is often
limited by the adhesion between the layers. Experimental observations and measurements
of interfacial properties in ceramic-polymer/metal systems are described. Spherical
shaped indenters have been used to induce delaminations at the interface. The load for
initial delamination is used to measure the interfacial strength, while the load-crack
length relationships are used to estimate interfacial toughness. In the ceramic-polymer
system used for dielectric protection, surface chemical modifications led to enhanced
interfacial toughness, and various modelling approaches needed to properly characterize
the system are discussed. In the ceramic-metal cases described, the interfacial
microstructure changes during processing or use (high temperature sealing, and solid-
oxide fuel cell applications). Here we will highlight the reasons why multi-scale models
are necessary for accurate descriptions of performance, and how experimental techniques
can help in building these models.
Sandia National Laboratories is a multi-program laboratory managed and operated by
Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the
U.S. Department of Energy's National Nuclear Security Administration under contract
DE-AC04-94AL85000
 Decoding the chemomechanics of friction and scratch in complex granular hydrated oxides

Soroosh jalilvand1, Rouzbeh Shahsavari1,2

1
Department of Civil and Environmental Engineering, Rice University, Houston, TX 77005
2
Department of Material Science and NanoEngineering, Rice University, Houston, TX 77005

Calcium-Silicate-Hydrate (C-S-H) is a structurally complex material that is the primary

binding phase of all Portland cement concrete materials. C-S-H is typically considered
as an assemblage of discrete nanoscale particles whose interactions are governed by
nano-scale friction and interfacial cohesion. Despite the critical role of interparticle forces
on mechanics of granular materials such as C-S-H, there is currently no unified
understanding on the atomistic mechanisms governing the nanoscale friction, scratch and
cohesion in C-S-H; a lack of knowledge, which presents a major barrier to decode the
interplay between chemistry, structure and mechanics of various C-S-H systems. Here, we
develop a molecular dynamics framework, coupled to a set of re-defined classical
continuum relations, aimed at filling this gap. The normal force is calculated by relaxation of
two C-S-H particles at various distances and
crystallographic orientations. The effect of
watery environment is studied by immersing
the particles in water molecules, and the
friction and scratch forces are calculated by
sliding the tip over the substrate in various
directions. Our study identifies the distinct
contribution of van der Waals and electro-
static forces to the interfacial behavior of C-
S-H particles. While the electrostatic forces
govern the interaction of particles at short and
large distances, the van der Waals forces are
responsible for variations in the normal force
at intermediate distances. We find that normal
force varies sub-linearly with nanoscale
contact area between the tip and substrate,
while the friction force shows a linear trend.
Our results demonstrate to what extent
Figure 1. Typical scratch test in C-S-H (top).
atoms of particular type contribute to
Classical fracture size-effect of quasi-brittle materials
does not hold for nanoscale scratch test in C-S-H.
the total interfacial forces. Due to large
electronegativity, Si atoms (vs Ca, O and H)
contribute most to the friction and normal forces. Finally, we probe the effect of pile-up and
show that the nanoscale scratch test does not follow the classical laws of size-effect in
fracture of quasi-brittle materials (Figure 1). Our findings, for the first time, provide an
“atomistic lens” on nanoscale friction and contact phenomena in particulate C-S-H systems.
This has a significant impact on fundamental understating of C-S-H and modulation of its
behavior, thus impacting the mechanics of granular cementitious materials.
[1] Shahsavari R., Buehler M., Pellenq R., Ulm F.-J., Journal of American Ceramic Society, 92, 2323
(2009).
[2] Pellenq R., Kushima A., Shahsavari R., Van Vliet K., Buehler M., Yip S., Ulm F.-J., Proceedings of
National Academy of Sciences, 09021180106, 1 (2009)